Re: protocol-27 comments

[Hallvard B Furuseth <h.b.furuseth@usit.uio.no>]

Jim Sermersheim writes:
>>>> Hallvard B Furuseth <h.b.furuseth@usit.uio.no> 10/29/04 6:40:42 AM
>> My point is, we need something which says that the _server_ may either
>> perform the modification or not.
> 
> I don't think it really matters, but I can change it to:
> "Whether the modification is applied or not is indeterminate if the
> Modify Response is not sent or received (e.g. the LDA exchange is
> terminated or the Modify Operation is abandoned)."
>  
> Note that the server may have sent the response, but it was never
> received due to a communincations problem, so it's still indeterminate
> from the client's point of view.
> 
> Of course this is true of any update operation, and maybe it belongs
> somewhere else.

Yes, that's true.  We have already got section 3.1 (Operation and LDAP
Exchange Relationship), and I'd missed that 4.6 already says:

   - changes: A list of modifications to be performed on the entry. The 
     entire list of modifications MUST be performed in the order they 
     are listed as a single atomic operation.

Maybe all the text from "Due to the requirement for atomicity..." in the
section we are discussing be deleted in favor of adding this after the
text above:

     The server MUST perform either all the changes or (e.g. if the
     operation fails or is abandoned) none of the changes.

-- 
Hallvard